Palo Alto CEO Calls AI Slowdown Unrealistic, Says Extinction Risk Is Below 10%

1 articles · Updated · CNBC · Sep 24

Nikesh Arora said slowing AI development is unrealistic because frontier work will continue somewhere, making broad industry “pacing” hard to enforce.
Arora argued developers should police themselves instead—testing models, withholding unsafe products and building guardrails—rather than seeking regulatory sympathy or liability protection.
The Palo Alto Networks chief also rejected claims of a 10% chance AI could wipe out humanity, calling that probability “extremely small.”
His comments push back on Anthropic CEO Dario Amodei’s call to slow capability gains, a view also backed by Sam Altman, Elon Musk and Demis Hassabis.
The split comes as Washington weighs AI oversight and even limited U.S.-China coordination, though Arora said the U.S. first needs a clearer domestic consensus.